Problem
Research questions and friction points this paper is trying to address.
terminal task synthesis
cross-artifact consistency
information preservation
executable state
FACET: Preserving Source Intent and Executable State in Terminal Task Synthesis
π€ AI Summary
ζ¬ζζεΊFACETζ‘ζΆοΌιθΏδΏζζΊζεΎεζ§θ‘ηΆζηδΈθ΄ζ§ζ₯解ε³ι«θ΄¨ιη»η«―δ»»ε‘εζιΎι’οΌζι«ηη£ζηγ
π Abstract
Training terminal agents requires scalable executable supervision, yet synthesizing high-quality terminal tasks remains challenging. Each task couples an instruction, an initialized environment, a reference solution, and an executable verifier; if these artifacts are generated from inconsistent assumptions, the resulting task may be unsolvable or incorrectly evaluated. Meanwhile, multi-stage synthesis can discard the goals, dependencies, state transitions, and procedural constraints encoded in the original sources. We present FACET (Fine-grained Agentic Construction of Executable Tasks), a framework that addresses both information preservation and cross-artifact consistency. FACET reconstructs related agent skills into coherent, information-rich scenarios, then realizes and repairs the execution environment before generating the final task artifacts. The resulting container state serves as shared grounding for the instruction, solution, and verifier, while execution-based validation and targeted repair correct artifact-specific failures without unnecessarily regenerating valid components. FACET produces complex terminal tasks with dense executable checks, and successful trajectories collected from these tasks provide effective, data-efficient supervision. Fine-tuning models across multiple scales consistently improves performance on Terminal-Bench 2.1, while analyses of alternative generation schemes support the importance of environment-grounded construction for task validity and solution-verifier alignment. These results establish source-intent preservation and shared executable-state grounding as key principles for scalable terminal-task synthesis.
Innovation
Methods, ideas, or system contributions that make the work stand out.
executable supervision
cross-artifact consistency
environment-grounded construction
π Similar Papers
No similar papers found.
